Speaker
Description
The Fermi Large Area Telescope (LAT) detected more than 7000 𝛾-ray sources in 14 years of operation which are collected in the 4FGL-DR4 catalogue. About a third of these sources are still unassociated with counterparts in other wavelength and approximately one fifth are associated with blazar of unknown type, as their classification as either BL Lac type blazars or flat spectrum radio quasars is still unclear. We developed a machine learning method based on artificial neural networks trained with the 4FGL-DR4 identified sources multi-wavelength data. We used this method to classify blazar of unknown type as possible BL Lac type blazars or flat spectrum radio quasar. Then we performed a three-category classification of the 4FGL-DR4 catalogue sources using the same method to characterize them on the likelihood of being a pulsar, a BL Lac type blazar or a flat spectrum radio quasar. We used the classification results to propose a list of possible unidentified 𝛾-ray sources multi-wavelength counterparts.